East Kirk of St Nicholas dig November 2008
Judith Stones, Keeper of Archaeology, writes…
For this latest edition of the ‘blog’, I’m delighted to introduce a report by Carmen Cuenca-Garcia about her recent ground-penetrating radar survey in the transepts and crossing of St Nicholas. If you want to remind yourself of the apse, which features in her story and formed the east end of the earliest church structure uncovered during the dig, you might like to look back at earlier editions of this diary – especially Weeks 44 and 45 (November/December 2006) and January 2007.
And now over to Carmen herself……
I took part in the excavation in 2006 which included the discovery of an apsidal church which may date to the 11th or early 12th century. This prompted my interest in the transepts and so when I had to choose a subject for the dissertation during my MSc in Archaeological Prospection course at the University of Bradford, I decided to revisit St Nicholas Church. Previous archaeological excavation carried out by John Hunter, now Birmingham University’s Professor Hunter, in 1974, in the north transept (Collison’s Aisle / St John’s Chapel) and part of the crossing did not identify the presence of any structure predating the 12th century. In the light of these facts I decided that further investigation in the area underneath the south transept and the crossing was potentially interesting (figure 1).
Figure 1: 1847-1895 St Nicholas Kirk. Note the survey area highlighted in red (adapted from Google Earth 2008)
Since additional intrusive work could not be undertaken, a ground-penetrating radar (GPR) survey was considered the most appropriate technique to employ as it is fully non-invasive. GPR involves the transmission of high frequency radar pulses from a surface antenna into the ground. It measures the time taken for this energy to be transmitted and reflected by buried remains and soil deposits and returned back to the surface. The radar reflections are measured, recorded and displayed in real-time by dragging the antennae along parallel lines.
The flagstone floor of the transepts allowed a good contact of the antennae to the ground. This is important to avoid a loss of the radar energy penetrating into the ground and the introduction of air waves in the data and other spurious signals from features located above ground.
Furthermore, good contrast conditions were anticipated from the combination of dry sandy soils containing potential big stone structures, which indicated that further remains could be detected during the survey. The known depth of burial of potential archaeological features was expected to be between about 1-1.75m (figure 2).
Figure 2: View of the apse (bottom of photograph) showing the approximate level of the floor in the crossing and transept and the estimated depth of structural targets
All these aspects provided an ideal scenario to carry out a GPR survey.
After a pilot survey carried out to test the quality of the signal in the area and to evaluate the limitations of noise problems (e.g., the potential noise caused by the backfill of the previous archaeological investigation and building works carried out in the area), a more detailed survey was undertaken.
Prior to the GPR survey, the transepts, crossing and the church wall found in the excavation in the East Kirk were PenMap surveyed by Eline Schotsmans (figure 3, centre). The aim was to create an accurate plan to locate the survey grid and to correlate the apse and other surface features (e.g. tombstones) with the GPR data.
Figure 3: Eline Schotsmans and her PenMap survey team : Jack Dunbar & Iain Mulkerrin
Two grids were set up in an area of c.16m x 5.75m, covering the crossing and the south transept (figure 4).
Figure 4 : PenMap plan showing the location of the grids, surface features and early apsidal church ; dashed rectangle indicates the approximately outline of the 1974 archaeological evaluation
And now for the science!
The GPR system, a PulseEKKO 1000 unit from Sensors&Software (figure 4), was equipped with a 450 MHz frequency antenna. The goal was to map structures and other archaeological features in detail. Subsequently, a 225 MHz frequency antenna was also used to investigate the existence of archaeological features located at greater depths.
Figure 5 : PulseEKKO system (Eline Schotsmans)
A total of 60 profiles were acquired in parallel lines (figure 6).
Figure 6 : PenMap plan showing the location of the GPR survey lines
The 450MHz survey had a general profile spacing of 0.25m in N-S direction. Several perpendicular profiles were acquired in order to cross-correlate potential anomalies. The 225MHz survey was composed of 0.5m profile spacing in N-S direction.
The GPR survey team was formed by Alison Cameron, Hilary Hinton, Aidan Mulkerrin, Neil Paterson, Abeer Ralston, Laraine Selbie, Marlene Weston, Johan Yorston, and other volunteers and archaeologists from the dig (figure 7 to 10). About 150 people visited the church during survey and found out more about our work.
Figure 7: 450MhZ antenna survey in Drum’s Aisle. From left to right : St Nicholas Kirk members, Alison Cameron, Neil Paterson, Carmen Cuenca-Garcia, Hilary Hinton & Johan Yorston
Figure 8: Changing the antennae and 225MhZ antenna survey in Drum’s Aisle. From left to right : Carmen Cuenca-Garcia, Aidan Mulkerrin & Johan Yorston
Figure 9: 450MhZ antenna survey in the crossing. From left to right : Marlene Welston, Laraine Selbie, Johan Yorston, Abeer Ralston & Alison Cameron
Figure 10: Some of the team observing interesting reflections as the computer displays the GPR data in real-time
The raw GPR datasets were particularly difficult to interpret for two main reasons. Firstly, the mixed backfill of prior archaeological investigation and building works and secondly, the complexity of the archaeological remains expected inside a church (i.e. burials and other archaeological features cutting through earlier ones). The latter masked coherent geophysical responses with noise and attenuated the signal. The 2D processing of the data and its visualization as time-slices improved the datasets, facilitating the interpretation of the anomalies (figure 11). The software used for the processing was ReflexW.
Figure 11: Examples of GPR profile sections. Raw data (a) & processed data (b)
Since one of the main purposes of GPR surveys is to accurately map reflections in three dimensions, or time-slices (figure 12, 13 & 15), it is necessary to convert the two-way travel time scale (ns) into a depth scale (m), hence the GPR wave velocities must be determined. In this project the velocity was determined by the recording of a Common midpoint (CMP) gather and from hyperbola-fitting methods. A constant velocity of 0.115 m/ns-1 was assumed for the time-to-depth conversion of the time-slices.
Figure 12 : Time-slices from Grid 1 & 2, summarizing the main anomalies detected at a depth of 0.57m. Note the dashed red line (below A1/A2) indicates the approximately distance of 15 feet where the original south gable wall used to be
Figure 13 : Time-slices from Grid 1 & 2, showing potential geological anomalies (dotted line) and slight remains of the possible early church (A6) at a depth of 0.92 m
Figure 14: South facing view of Hunter’s evaluation
The results showed structural anomalies related to the foundations of the existing 12th century south-east pier (A6, figure 12) and some potential 11th or 12th-century building remains, including part of a surface (A6, figure 13). This surface might be the mortar floor associated with the apsidal church or an undated surface recorded in an archaeological evaluation carried out in the crossing in 1974.
Due to the high level of disturbance produced in the crossing area by the building works carried out in the church after the Reformation, the impact of the Great Fire of 1874, and the archaeological evaluation carried out in the area in 1975, the fragmentary reflections observed underlying the pier foundations (A6) and extending westwards (figure 13) might indicate the remains of an early church, potentially truncated by these building works.
Further anomalies might represent the remains of the original end of the south transept before it was extended southwards (A1 & A2, figure 12). Other high amplitude areas such as A3 (figure 12) might be produced by the debris accumulated after dismantling the original south wall of the transept.
A possible stone-lined grave (A4, figure 12) and a potential grave burial were also detected.
A large anomaly (A5, figure 12) in the centre of the crossing could be the platform used for the works to rebuild the tower after the great fire in 1874 (figure 14).
Other high amplitude reflections such as A7 and A8 (figure 12) may be clutter produced by isolated strong reflectors such as boulders, debris or metallic objects buried in the area after building works.
The survey also mapped the exact position of the evaluation trench excavated in 1974 and geological deposits (figure 15).
I will present further information about the survey and results during a talk to be held on the 3rd December at the Maritime Museum, Aberdeen. The talks are at 12.30 and 2pm and will last about 30 minutes. Booking essential: phone 01224 337714 to book a place.
- For further details, please contact judiths@aberdeencity.gov.uk
Back to the Excavation at Kirk of St Nicholas - Details home page.
